The development and maintenance of a tissue such as the retina is contingent upon the proper regulation of cell proliferation, cell survival and appropriate adhesive interactions between cells and their neighbors. While great advances have been made in our understanding of the genetic regulation of many of these processes, studies in Drosophila imaginal discs have highlighted the existence of modes of cell-cell interactions that we still do not understand. For instance, studies of the phenomenon of cell competition have demonstrated that cells are capable of assessing the fitness of their neighbors and subsequently influencing their proliferation or survival. We also know that cells appear to preferentially adhere to other cells from the same portion of the imaginal disc while avoiding interactions with cells from other parts of the same primordium. This preference may be absolute such as the clear separation of anterior and posterior cells in leg and wing discs or relative, such as the tendency of clonal boundaries to fall along the equator in the developing Drosophila eye. While both cell competition and compartmentalization were discovered almost forty years ago, we still do not have a clear understanding of the mechanistic basis of either phenomenon. It is therefore likely that much remains to be discovered about the ways in which cells interact with the neighbors in tissues such as the developing retina. Mosaic screens in the Drosophila eye have been used to discover many genes that function in regulating cell proliferation, cell fate specification and photoreceptor differentiation. Importantly, the phenotypes elicited by mutations in some of these genes are only obvious in a situation where the disc is composed to two different types of cells: wild type and mutant. Under these conditions, changes in cell proliferation manifest as a change in the relative representation of the two populations of cells without necessarily affecting eye size and changes in cell-cell affinity can result in the formation of smooth, rather than irregular boundaries between the two populations. While screens based on RNA-mediated interference (RNAi) have been extremely effective in identifying genes in a number of biological processes, it is currently difficult to us this approach in the context of a mosaic screen. We have developed a new method, CoinFLP, which allows us to conduct mosaic screens in the Drosophila eye using RNAi and have validated this approach with a pilot screen that has already identified several genes not previously implicated in eye development. We propose to use this approach to identify and characterize genes that regulate cell proliferation and cell affinity in the developing Drosophila eye.